PC reboots when video card is under stress

Sounds like the OP's problem is a bad power supply. Swap it out and test it. I'm honestly a bit surprised that it's restarting instead of BSOD.

jojo69 said:
so guys, let me slightly derail as I have a similar thing going on

2 6950s in crossfire, original issue Corsair AX1200

after some time under load the screens would go black, computer appeared to still be running, this happened several times. I tried reseating everything related to the GPUs and blowing out the dust, still happened.

Pull the #2 card and now everything is fine...except of course my framerates are down


#2 card or PSU??
Click to expand...
I'm willing to bet that this is a heat related issue. Since you already pulled the cards and cleaned out the dust, make sure the fans are working and ramping up to speed correctly. Then while running a game, check that the temps of each card aren't unreasonable high. I had similar issues with my trifired 7950's. Two cards were fine, but the top card was being cooked and couldn't cool itself even at 100% fan speed.
 
NukeULater said:
Sounds like the OP's problem is a bad power supply. Swap it out and test it. I'm honestly a bit surprised that it's restarting instead of BSOD.

I'm willing to bet that this is a heat related issue. Since you already pulled the cards and cleaned out the dust, make sure the fans are working and ramping up to speed correctly. Then while running a game, check that the temps of each card aren't unreasonable high. I had similar issues with my trifired 7950's. Two cards were fine, but the top card was being cooked and couldn't cool itself even at 100% fan speed.
Click to expand...
image.jpg

You didn't read the hole thread apparently because as seen here were waiting for op to double check his rails since they look pretty bad
 
yeah what prime said...
instant reboots are not usually heat related. heat takes time to build, power loss is instant....

edit; oh youre replying to jojo who thread jacked...
 
primetime said:
You didn't read the hole thread apparently because as seen here were waiting for op to double check his rails since they look pretty bad
Click to expand...
Those voltages are terrible. Apparently you quickly read my post, since I quoted and was replying to Jojo. That definitely seems like an overheating issue, unlike the thread creator's problem.
 
Jojo needs his own thread instead of jacking this one. The issues are totally different.

OP, you still alive, have any luck?
 
NukeULater said:
Those voltages are terrible. Apparently you quickly read my post, since I quoted and was replying to Jojo. That definitely seems like an overheating issue, unlike the thread creator's problem.
Click to expand...
lol yea my bad.....I read most of your post:)
 
I had a very similar crash before when using two cards under stress. It turned out the PSU was too weak to handle the load.
 
Another thing to consider is that the longer you let this power condition continue, the more chance of damaging other components in your system. That's no joke. Signs point to bulging or leaking capacitors in your power supply which has resulted in decreased capacity on one or multiple rails. You may even be able to see them if you take the power supply out of the case and look inside with a flashlight. If you can find a damaged capacitor or capacitors, that's all you need to start an RMA.
 
I'm back. I couldn't test the PSU because I was busy with other stuff, and now that I have made some time to test it, I found that all the cables in this f*cking PSU are black, so I don't even know how to do the "paper clip trick" because I don't know which is the green wire, not to say how to test the PCI-E cables which are also black. So I'm still fighting with that.
 
It the psu. Just had the same problem recently and had to Rma my ax1200i. Psu would just reboot at random times. My hx850 wasn't enough to power sig rig and would shut off also. Straight up had to wait for the replacement to game.
 
Amadeo said:
I'm back. I couldn't test the PSU because I was busy with other stuff, and now that I have made some time to test it, I found that all the cables in this f*cking PSU are black, so I don't even know how to do the "paper clip trick" because I don't know which is the green wire, not to say how to test the PCI-E cables which are also black. So I'm still fighting with that.
Click to expand...
paper clip? I guess you already replaced the power supply then?
 
Amadeo said:
I'm back. I couldn't test the PSU because I was busy with other stuff, and now that I have made some time to test it, I found that all the cables in this f*cking PSU are black, so I don't even know how to do the "paper clip trick" because I don't know which is the green wire, not to say how to test the PCI-E cables which are also black. So I'm still fighting with that.
Click to expand...

You could just locate its position aka count where it is.
http://www.smps.us/atx-connector-20-24pin.jpeg
 
Now, instead of measuring the voltages with the multimeter, I could use nvidia-smi to get that information, right? And in principle I suppose it is reliable information, the same that I would measure with the multimeter.

For example, nvidia -q -d POWER gives me the following:

Timestamp : Sat May 7 03:37:13 2016
Driver Version : 361.42

Attached GPUs : 1
GPU 0000:01:00.0
Power Readings
Power Management : Supported
Power Draw : 17.34 W
Power Limit : 300.00 W
Default Power Limit : 300.00 W
Enforced Power Limit : 300.00 W
Min Power Limit : 150.00 W
Max Power Limit : 390.00 W
Power Samples
Duration : 494.71 sec
Number of Samples : 119
Max : 72.50 W
Min : 16.60 W
Avg : 17.52 W
Click to expand...
 
Amadeo said:
Now, instead of measuring the voltages with the multimeter, I could use nvidia-smi to get that information, right? And in principle I suppose it is reliable information, the same that I would measure with the multimeter.

For example, nvidia -q -d POWER gives me the following:
Click to expand...
lol in all honesty your making a very simple procedure extremely complicated in my mind. Just grab an unused molex connector, put the meter leads in the end and read what the meter tells you lol..you dont have to see any colors to do this
molex-fan-connector.jpg

The grounds are the 2 middle wires and on each end will be 12v and 5v wires. One of your test leads goes on the ground and the other on the rail your wanting to test.

You can find different software methods of reading the voltage rails all day long and it doesn't hurt a thing, but your missing the point of using a meter to check the measurements. Software measurements dont work correctly on all pc's, and thats the point of using a meter.
 
Last edited:
You must log in or register to reply here.
Back
Top